Talk:KOI character encodings

Does KOI8-CS for Czech and Slovakian really exist? Both languages are written with the latin alphabet. I found no nothing about KOI8-CS on the Web. I'll remove it in a few days unless somebody provides proof of their existance.

91.41.107.190 (talk) 14:41, 1 March 2009 (UTC)

(I know this is old, but still) I, too, was surprised, however it seems that they really do exist. I've added the respective references to the article. 78.50.190.53 (talk) 01:14, 19 April 2011 (UTC)

KOI8-E

The following text was in the article:

*[[KOI8-E]] / [[KOI8-CYRILLIC]] (ISO-IR-153,<ref>[https://www.itscj.ipsj.or.jp/iso-ir/153.pdf ISO-IR-153 (1 December 1989)]</ref> ST SEV 358-88, traditional Cyrillic order, only for Russian<!-- Please, do not cite GOST 19768-74 which only defined the original KOI-8 with АБЦД..., this is a persistent error repeated here and there.-->).<ref name="Cruz_2010_Kermit"/>

Now, the Kermit reference only mentions GOST 19768-74 (noted in the existing comment as defining the original KOI-8), not ST SEV 358-88 or ISO-IR-153. Moreover, the IANA list KOI8-E as meaning ISO-IR-111. (As a sidenote, KOI8-E and KOI8-CYRILLIC redirect to ISO-IR-111.)

The applicability of "KOI8-E" to ISO-IR-153 specifically is therefore unsourced, and the applicability of "KOI8-E" to ISO-IR-111 in at least some contexts is therefore sourced (IANA).

Add to this that ISO-IR-153 does not follow the KOI layout (it's a subset of ISO-8859-5), so calling it a KOI encoding just because it encodes the Russian alphabet seems a bit of a stretch. In any case, I'm removing it until / unless someone can adequately source ISO-IR-153 (rather than GOST 19768-74 or ISO-IR-111) being referred to as KOI8-E in any notable context. -- HarJIT (talk) 00:46, 13 April 2018 (UTC)

---

The historical preamble to ECMA 113-1988 seems to explain what is going on.

In brief: ECMA 113-1988, noted as being equivalent to ISO 8859-5, is noted to be in conformance with GOST 19768-1987, which is presumably what ISO-IR-153 (being an ISO 8859-5 subset) actually means to cite when it incorrectly lists GOST 19768-74 as one of its source standards. This incompatible change to ECMA 113 is noted as being a result of the GOST standard being revised, i.e. that ECMA 113-1986 (ISO-IR-111) would have been in conformance with GOST 19768-74 / KOI-8, which it is.

I've added an explanation of this to the ISO-IR-111 page (to which KOI8-E was already a redirect, as noted above). -- HarJIT (talk) 13:24, 13 April 2018 (UTC)